EP3649786A1

EP3649786A1 - Method for encoding and decoding images, encoding and decoding device, and corresponding computer programs

Info

Publication number: EP3649786A1
Application number: EP18752538.1A
Authority: EP
Inventors: Joël JUNG; Félix Henry; Bappaditya Ray
Original assignee: Orange SA
Current assignee: Orange SA
Priority date: 2017-07-05
Filing date: 2018-06-28
Publication date: 2020-05-13
Also published as: US11284085B2; WO2019008253A1; US20200128251A1; KR102701711B1; FR3068558A1; KR20200022492A; JP2020526119A; CN111034202B; CN111034202A

Abstract

The invention relates to the encoding of an image (ICj) divided into blocks, said image containing separate first and second zones (Z1, Z2). The encoding is characterised in that it implements the following, for at least one current block (B_u) of the image: determining (C4) to which of the first and second zones the current block pertains; if the current block pertains to the first zone (Z1), encoding (C5a)) the current block by means of a first encoding method (MC1); if the current block pertains to the second zone (Z2), encoding (C5b)) the current block by means of a second encoding method (MC2) comprising the following: from the position of the current block in the second zone, identifying (C51b)) a previously encoded, then decoded block (B_ref) located in the first zone of the image; and reproducing the value of at least one encoding parameter (PRC₁) associated with the identified block.

Description

METHOD FOR ENCODING AND DECODING IMAGES, DEVICE FOR ENCODING AND DECODING AND COMPUTER PROGRAMS

CORRESPONDENTS Field of the invention

The present invention relates generally to the field of image processing, and more specifically to coding and decoding of digital image parameters, whether these digital images are fixed or part of a digital image sequence.

The coding / decoding of such image parameters applies in particular to images originating from at least one video sequence comprising:

images coming from the same camera and succeeding each other temporally (coding / decoding of 2D type),

images from different cameras oriented according to different views (coding / decoding of 3D type),

corresponding texture and depth components (3D type coding / decoding),

images obtained by projection of a 360 ° video,

- etc.

The present invention applies similarly to the coding / decoding of 2D or 3D type image parameters.

In particular, but not exclusively, the invention can be applied to the video coding implemented in the current video codecs AVC (abbreviation of "Advanced Video Coding") and HEVC (abbreviation of "High Efficiency Video Coding") and their extensions (MVC, 3D-AVC, MV-HEVC, 3D-HEVC, etc.), and the corresponding decoding.

Background of the invention

Current video encoders (MPEG, H.264, HEVC, ...) use a block representation of the video sequence. The images are cut into blocks, which can be re-cut, for example recursively as in the HEVC standard. For a current block to be coded, the image parameters associated with this block are coded in the form of bits by means of a coding method adapted implemented by an encoder, such as for example an entropic coder whose purpose is to code these parameters without loss.

Such parameters are for example:

the residual prediction coefficients of the pixels of the current block,

the mode of prediction of the current block (Intra prediction, Inter prediction, prediction by default carrying out a prediction for which no information is transmitted to the decoder ("in English" skip ")),

information specifying the type of prediction of the current block (orientation, reference image, etc.),

- the type of division of the current block,

- the movement information of the current block if necessary, - etc.

The bits obtained after entropy coding are written in a data signal which is intended to be transmitted to the decoder.

Once the coded data signal has been received by the decoder, the decoding is done image by image, and for each image, block by block. For each block, the bits representative of the image parameters associated with the block are read and then decoded using a decoding method implemented by a decoder.

For each type or image format considered, a specific encoding is implemented. For example, the AVC and HEVC coders / decoders are suitable for coding / decoding 2D images coming from the same camera and succeeding one another temporally. Such encoders / decoders are also suitable for coding / decoding:

images obtained by projection of a stereo video and each comprising two representative views of the same scene, which are intended to be viewed respectively by the left eye and the right eye of the user,

images obtained by two-dimensional projection of a 360 ° stereo video, etc. According to another example, the 3D-HEVC coders / decoders are suitable for coding / decoding 3D images, such as 3D images from different cameras oriented in different views, corresponding texture and depth components, images derived from different images. a 360 ° mono video, etc.

By thus proposing a very specific type of coding / decoding for a given format or image type, the coding / decoding performances obtained by the current coders / decoders are not satisfactory.

The invention therefore aims at an encoder / decoder that proposes to use two different coding / decoding methods for a current image derived from a video content of a given type or format, while making it possible to optimize the coding / coding performance. decoding of the current image.

Object and summary of the invention

One of the aims of the invention is to overcome disadvantages of the state of the art mentioned above.

For this purpose, an object of the present invention relates to a method of encoding a block-cut image, said image containing first and second distinct areas.

Such a coding method is remarkable in that it implements the following, for at least one current block of the image:

determining which of the first and second zones belongs to the current block,

if the current block belongs to the first zone, coding the current block using a first coding method,

if the current block belongs to the second zone, coding the current block using a second coding method which includes the following:

starting from the position of the current block in the second zone, identifying a previously coded block, then decoded, situated in the first zone of the image,

- copy the value of at least one coding parameter associated with the identified block. Such an arrangement makes it possible, within the same encoder, to select a coding method that is most adapted to the way in which the scene represented by the image was initially captured.

For example, if the current image is a two-dimensional image, it is decided:

in the case where this image comes from a video comprising temporally succeeding 2D images, to select a first coding method implemented in the encoder, such as a conventional method of AVC or HEVC type;

in the case where this image has been obtained by projection of a captured stereo video according to a plurality of viewing angles for example covering an angle of 360 °, this image being formed of two image zones corresponding to the eye left and the right eye of the user, to select in the encoder is the first encoding method mentioned above, or a second coding method depending on whether the current block is in the first zone or in the second zone of the 'picture.

Thanks to the invention, it is thus possible to propose an adaptive coding of video contents that takes into account the format of the latter. In addition, for certain types of content format, the invention advantageously makes it possible to apply an adaptive coding to the current image according to the zone of the image to be encoded at the current time, at least one current block of a first zone being coded according to a first coding method, and at least one current block of a second zone of this image being coded according to a second coding method.

The second coding method advantageously makes it possible to take advantage of the spatial correlations that exist between the second zone and the first zone of the current image, the value of at least one coding parameter of a block located in the first zone being copied for a current block of the second zone. It is thus not necessary to code the value of such an encoding parameter for the current block. This results in a more efficient image coding because it is more precise, less complex and less expensive in terms of throughput. According to a particular embodiment, the second coding method is applied to all the blocks of the second zone of the current image.

Such an arrangement makes it possible to optimize the reduction of the complexity and the reduction of the encoding bit rate implemented at the encoder.

According to another particular embodiment, the coding method implements the following:

coding said at least one current block of the second zone of the image by means of the first coding method,

selecting the first coding method or the second coding method in accordance with a predetermined coding performance criterion,

coding information representative of said selection.

In view of the fact that the first and second coding methods are in competition, although the second coding method was first applied to the current block encoder, the coding method is made more flexible. improves coding efficiency.

According to yet another particular embodiment, the first and second zones of the image have the same shape, the first zone being situated above the second zone and separated from the latter by a horizontal boundary extending in the middle of the second zone. the image.

Such a spatial arrangement of the first and second zones of the current image makes it possible to optimize the coding performances, by precisely pre-defining the location of a block already coded and then decoded from the first zone of the image, as a function of the position. of the current block of the second zone.

The various embodiments or embodiments mentioned above may be added independently or in combination with each other, to the operations implemented during the coding method as defined above.

Correlatively, the invention relates to a device for encoding at least one image cut into blocks, said image containing first and second distinct areas. Such a coding device is remarkable in that it comprises a processing circuit which is arranged to implement the following, for at least one current block of the image:

determining which of the first and second zones belongs to the current block,

- copy the value of at least one coding parameter associated with the identified block.

Correspondingly, the invention also relates to a method of decoding a data signal representative of a block-cut image that has been encoded, said at least one image containing first and second distinct areas.

Such a decoding method is remarkable in that it implements the following, for at least one current block to decode the image:

determining which of the first and second zones belongs to the current block,

if the current block belongs to the first zone, decoding the current block using a first decoding method,

if the current block belongs to the second zone, decoding the current block using a second decoding method which comprises the following:

starting from the position of the current block in the second zone, identifying a previously decoded block located in the first zone of the image,

assigning to the current block at least one decoding parameter associated with said identified block. According to a particular embodiment, the second decoding method is applied to all the blocks of the second zone of the current image.

According to another particular embodiment, the second decoding method is applied to the current block of the second zone if, for the current block, is read in the data signal selection information of the second decoding method, the first method decoding being applied to the current block of the second zone if, for the current block, is read in the data signal selection information of said first decoding method.

The various embodiments or embodiments mentioned above may be added independently or in combination with each other, to the operations implemented during the decoding method as defined above.

Correlatively, the invention relates to a device for decoding a data signal representative of a block-cut image that has been encoded, said at least one image containing first and second distinct areas.

Such a decoding device is characterized in that it comprises a processing circuit which is arranged to implement the following, for at least one current block to decode the image:

determining which of the first and second zones belongs to the current block,

if the current block belongs to the second zone, decoding the current block using a second decoding method which comprises the following: starting from the position of the current block in the second zone, identifying a previously decoded block located in the first zone of the image,

assigning to the current block at least one decoding parameter associated with said identified block.

The invention also relates to a computer program comprising instructions for implementing one of the coding and decoding methods according to the invention, when it is executed on a computer.

Such a program can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any another desirable form.

Still another object of the invention is directed to a computer readable recording medium, and including computer program instructions as mentioned above.

The recording medium may be any entity or device capable of storing the program. For example, the medium may comprise a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording means, a digital recording means, for example a key USB or a hard drive.

On the other hand, such a recording medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means. The program according to the invention can be downloaded in particular on an Internet type network.

Alternatively, such a recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute the method in question or to be used in the execution of the latter. Brief description of the drawings

Other features and advantages will appear on reading preferred embodiments described with reference to the figures in which:

FIG. 1A represents the progress of the coding method according to a first embodiment of the invention,

FIG. 1B represents the progress of the coding method according to a second embodiment of the invention,

FIG. 2A represents a coding device according to a first embodiment of the invention,

FIG. 2B represents a coding device according to a second embodiment of the invention,

FIGS. 3A to 3C each represent examples of distinct zones in the current image to be encoded or decoded,

FIG. 4 represents an example of a conventional coding method implemented in the coding methods of FIGS. 1A and 1B,

FIGS. 5A and 5B each represent two different examples of identifying a reference block in a current image, when applying to the current block a coding or decoding method according to the invention, as set out in FIG. implemented in the coding methods of FIGS. 1A and 1B or in the decoding methods of FIGS. 6A and 6B,

FIG. 6A represents the progress of the decoding method according to a first embodiment of the invention,

FIG. 6B represents the progress of the decoding method according to a second embodiment of the invention,

FIG. 7A represents a decoding device according to a first embodiment of the invention,

FIG. 7B represents a decoding device according to a second embodiment of the invention,

FIG. 8 represents an example of a conventional decoding method implemented in the decoding methods of FIGS. 6A and 6B. Detailed description of the coding part

A first embodiment of the invention will now be described, in which the coding method according to the invention is used to code an image or a sequence of images according to a bit stream close to that obtained by a coding implemented in an encoder according to any current or future video coding standards.

In this embodiment, the coding method according to the invention is for example implemented in a software or hardware way by modifications of such an encoder. The coding method according to the first embodiment of the invention is represented in the form of an algorithm comprising operations C1 to C6a) or C1 to C6b) as represented in FIG. 1A.

According to the first embodiment of the invention, the coding method is implemented in a coding device or coder CO shown in FIG. 2A.

As illustrated in FIG. 2A, the coder CO comprises a memory

MEM_C comprising a buffer memory MT_C, a processor PROC_C controlled by a computer program PG_C which implements the coding method according to the invention. At initialization, the code instructions of the computer program PG_C are for example loaded into a RAM, denoted MR_C, before being executed by the processor PROC_C.

The coding method represented in FIG. 1A applies to any current image IC _{j that is} fixed or part of a sequence of L images ld, IC _j , ..., IC _L (1 _{j j} _{L L} ) to code.

The current image IC _j comes from at least one video sequence comprising, by way of non-exhaustive examples:

corresponding texture and depth components, ie representative of the same scene (coding / decoding of 3D type),

images obtained by projection of a mono 360 ° video, images obtained by projection of a stereo video and each comprising at least two representative views of the same scene,

non-natural images of the "screen content" type, such as, for example, images obtained by screen video capture,

- etc.

With reference to FIG. 1A, C1 is, in a manner known per se, the cutting of a current image IC _j into a plurality of blocks B _; B ₂ , B _u , ..., B _s (1 < _u _S S). The partitioning is implemented by a partitioning device MP_C shown in FIG. 2A, which device is controlled by the processor PROC_C.

It should be noted that for the purposes of the invention, the term "block" means coding unit (coding unit). This last terminology is notably used in the standard HEVC "ISO / IEC / 23008-2 Recommendation ITU-T H.265 High Efficiency Video Coding (HEVC)".

In particular, such a coding unit groups together sets of pixels of rectangular or square shape, also called blocks, macroblocks, or sets of pixels having other geometrical shapes.

Said blocks B _; B ₂ , B _u , ..., B _s are intended to be coded according to a predetermined order of travel, which is for example of the lexicographic type. This means that the blocks are coded one after the other, from left to right.

Other types of course are of course possible. Thus, it is possible to cut the image IC _j into several subimages called slices and to independently apply a division of this type on each sub-image. It is also possible to code not a succession of lines, as explained above, but a succession of columns. It is also possible to browse the rows or columns one way or the other

According to one example, the blocks Bi, B ₂ , B _u , ..., B _s have a square shape and all contain K pixels, with K> 1. By way of non-exhaustive example, the blocks have a size of 64 × 64 pixels and / or 32 × 32 and / or 16 × 16 and / or 8 × 8 pixels.

Depending on the size of the image that is not necessarily a multiple of the size of the blocks, the last blocks on the left and the last blocks on the bottom may not be square. In an alternative embodiment, the blocks can be for example of rectangular size and / or not aligned with each other.

Optionally, as shown in dashed lines in FIG. 1A, C2 is coded for an activateStereoReuse syntax element that is associated with a characteristic of the current image IC _j .

The activateStereoReuse syntax element is a high-level syntax element of a video sequence that includes the current image IC _j . For this purpose, depending on the coding context, this element can be coded:

- at the beginning of the coding of each image of the video sequence, - or only once at the beginning of the coding of a sequence of images,

- or only once at the beginning of the coding of the video sequence. The activateStereoReuse syntax element is intended to indicate the type of current image format to be encoded. According to a preferred embodiment, the activateStereoReuse syntax element is coded at the value 1 if the current image to be encoded has been obtained by projection of a stereo video, 360 °, 180 ° or others, and the image current is composed of several views captured at the same time instant and arranged in the current image so as to form a single view (rectangle of pixels). A method of composing such an image uses, for example, the technique called "Frame Packing" (FP). On the other hand, the activateStereoReuse syntax element is coded to 0 if the current image to be encoded is of 2D type or has been obtained by projection of a mono video, 360 °, 180 ° or others.

The coding C2 is for example a entropic coding of type CABAC

("Context Adaptive Binary Arithmetic Coder" in English) or else an entropy coding of arithmetic type or Huffman. This coding is implemented by a coding device MC_C represented in FIG. 2A, which device is controlled by the processor PROC_C.

Such C2 coding is not necessary in the case where the CO encoder operates independently a distinction between:

a current image to be coded of 2D type or obtained by projection of a mono video, 360 °, 180 ° or others, and a current image to be encoded obtained by projection of a stereo video, 360 °, 180 ° or other and composed seltui a FP type technique.

In the remainder of the description, it is considered that the current image to be encoded has been obtained by projection of a stereo video, 360 °, 180 ° or other, and that the current image is composed of several views captured at the same time temporal and arranged in the current image so as to form a single view (rectangle of pixels).

With reference to FIG. 1A, the coder CO of FIG. 2A selects at C3 a current block to be coded B _u of the image IC _j .

In C4, the current block B _u of the image IC _{j is located,} for example by determining its coordinates with respect to the first pixel located at the top left of the image IC _j , and whose coordinates are ( 0.0). Such a location has the effect of determining whether the current block belongs to a first zone or a second zone of the current image IC _j , the first and second zones being distinct. According to the invention, the first and second zones are distinct, in the sense that they do not overlap.

The location C4 is implemented by a computing device CAL1_C as represented in FIG. 2A, which device is controlled by the processor PROC_C.

According to a first embodiment shown in FIG. 3A, which is the preferred embodiment, a current image IC _j comprises a first zone Z1 which extends over the upper half of the image and a second zone Z2 which extends over the lower half of the image. The zones Z1 and Z2 have the same shape and are separated from each other by a horizontal border FH extending in the middle of the image.

According to a second embodiment shown in FIG. 3B, a current image IC _j comprises a first zone Z1 which extends over the left half of the image and a second zone Z2 which extends over the right half of the image. picture. The zones Z1 and Z2 have the same shape and are separated from each other by a vertical border FV extending in the middle of the image.

According to a third embodiment shown in FIG. 3C, a current image IC _j comprises a first zone Z1 which extends over the first left top quarter of the image and a second zone Z2 which extends over the second upper left quarter of the image. The zones Z1 and Z2 have the same shape and are separated from each other by a horizontal border FH extending over a quarter of the image.

Other configurations are of course possible. For example, the zones Z1 and Z2 can be inverted. In addition, the zones Z1 and Z2 may have the same shape or not.

If the current block B _u belongs to the first zone Z1 of the image IC _j , with reference to FIG. 1A, the coding of the current block is carried out at C5a) using a first coding method MC1 . The first coding method MC1 is a conventional method, an example of which is illustrated in FIG. 4. The coding method MC1 is applied to any current block of the zone Z1.

With reference to FIG. 4, such a conventional encoding method MC1 implements a prediction C51 a) of the current block B _u by conventional prediction techniques, for example Intra and / or Inter and / or skip and / or Merge, etc. For this purpose, the current block B _u is predicted with respect to at least one predictor block according to a prediction mode belonging to one of the prediction techniques which have just been mentioned.

In a manner known per se, the current block B _u is predicted with respect to a plurality of candidate predictor blocks. Each of the candidate predictor blocks is a block of pixels that has already been encoded and decoded.

At the end of the prediction C51 a), an optimal predictor block BP _op t is obtained following a putting into competition of said predetermined prediction techniques, according to a predetermined coding performance criterion, for example by minimizing a flow criterion distortion well known to those skilled in the art. The BP _op t block is considered as an approximation of the current block B _u . The information relating to this prediction is intended to be written, in the form of syntax elements, in a signal or data stream to be transmitted to a decoder.

It is then conventionally carried out in C52a) to compare the data relating to the current block B _u with the data of the predictor block BP _op t. Such a comparison consists in calculating the difference between the predictor block obtained BP _opt and the current block B _u . A set of data, called residue block Br _u , is then obtained.

The operations C51 a) and C52a) are implemented by a predictive coding device PRED_C shown in FIG. 2A, which device is controlled by the processor PROC_C.

Referring again to FIG. 4, the data of the current residue block

Br _u are C53a) encoded in a conventional manner.

According to a nonlimiting exemplary embodiment, such an encoding C53a) implements the application C531 a) of a transform to the pixels of the current residue block Br _u .

In a manner known per se, depending on the context or the coding standard used, such a transform is for example a DCT (abbreviation of "Discrete Cosine Transform"), DST (abbreviation of "Discrete Sine Transform") ), type DWT (abbreviation of "Discrete Wavelet Transform") or type LT (abbreviation of "Lapped Transform"). These transforms are stored beforehand in an LTS1 list, in the MT_C buffer of the CO encoder of FIG. 2A.

At the end of the application of this transform, a current transformed data block Bt _u is obtained.

Such an operation is performed by a transform calculation device MTR_C, as represented in FIG. 2A, which device is controlled by the processor PROC_C.

The coding C53a) further implements a quantization C532a) of the data of the transformed block Bt _u according to a conventional quantization operation, such as, for example, a scalar or vector quantization. A block Bq _u of quantized coefficients is then obtained.

The quantization C532a) is implemented by a quantization device MQ_C as represented in FIG. 2A, which device is controlled by the processor PROC_C.

The transform calculation device MTR_C and the quantization device MQ_C are contained in a block encoding device

MCB_C shown in Figure 2A, which device is controlled by the processor

PROC C. The coding C53a) further implements a coding C533a) of the data of the block Bq _u of quantized coefficients. Coding C533a) is implemented by the coding device MC_C of FIG. 2A. At the end of the coding C53a), a set of coded data DC _U of the current block B _u is obtained.

With reference to FIG. 1A, it is proceeded at C6a) to construct a portion of signal F which conventionally contains:

the coded data DC _U obtained in C5a),

certain information encoded by the coder CO, such as, for example:

The type of prediction Inter, Intra, skip or Merge applied to the current block B _u , and if necessary, the prediction mode selected, the index of the predictor block obtained,

The type of partitioning of the current block B _u if the latter has been partitioned,

"The type of transform applied to the data of the current block B _u ,

- etc ....

According to the invention, the stream F optionally contains the value 0/1 of the activateStereoReuse syntax element, if the latter is coded at the image level.

The construction of the stream F is implemented by a data signal construction device MCF, as shown in FIG. 2A.

If, at the end of the location C4, the current block B _u belongs to the second zone Z2 of the image IC _j , the coding of the current block is carried out at C5b) using a second coding method MC2. According to the first embodiment, the second coding method MC2 is applied to any current block located in the second zone Z2.

According to the invention, with reference to FIG. 1A, C51 b) is used to identify a reference block B _ref which has been previously coded and then decoded, and which is located in the first zone Z1 of the current image IC _j .

The identification C51 b) is implemented by a computing device CAL2_C as represented in FIG. 2A, which device is controlled by the processor PROC_C. According to a preferred embodiment, if the current block which has been located in the second zone Z2 has its first pixel at the top left which has coordinates (x _u , y _u ) in the current image IC _j , then a block reference number B ' _re f is determined in the first zone Z1, as being the block whose first pixel pVef in the upper left has coordinates (x' _re f, yVet), such that x ' _re f = x _u and yVef = y _u -h / 2, where h is the height of the current image IC _j .

FIG. 5A represents an example of such a determination, in the case where the reference block B ' _ref which is determined does not overlap with other neighboring reference blocks in the first zone Z1 of the current image IC _j which is for example similar to that illustrated in Figure 3A. In Fig. 5A, the neighboring reference blocks are those represented with dots. According to such a configuration, the block B ' _ref is then considered as the block B _ref identified.

FIG. 5B represents another example of such a determination, in the case where the reference block B ' _ref which is determined overlaps with other neighboring reference blocks r1, r2, r3, r4 in the first zone Z1 of the current image IC _j which is for example similar to that illustrated in Figure 3A. According to such a configuration, according to an exemplary embodiment, it is determined which of the neighboring reference blocks r1, r2, r3, r4, the one with the most pixels in common with the block B ' _ref . In FIG. 5B, it is the reference block r4 which is then considered as the block B _ref identified.

Of course, there are other methods for selecting the neighboring reference block, when the block B ' _ref overlaps neighboring reference blocks in the first zone Z1 of the current image IC _j .

In another example, in the case of FIG. 5B, it could be determined which of the neighboring reference blocks r1, r2, r3, r4 is the one containing the center of the block B ' _ref .

According to yet another exemplary embodiment, the reference block B ' _ref is determined in the first zone Z1, as being the block which contains the coordinate pixel where (x _c , y _c ) are the coordinates of the center of the current block.

Since the reference block B _ref has been identified in the first zone Z1, with reference to FIG. 1A, at least one coding parameter P RCi associated with the reference block B _{ref is} read at C52b). Such a parameter encoding PRCi is stored in an LST2 list of buffer MT_C of the encoder of Figure 2A.

The reading C52b) is implemented by a reading device LEC_C as represented in FIG. 2A, which device is controlled by the processor PROC_C.

By way of non-exhaustive examples, the list LST2 contains a number K of coding parameters PRC-1, PRC ₂ ,..., PRC _K associated with the reference block B _ref identified, among which we find:

the type of prediction intra, inter, skip, merge, etc. chosen to predict the block B _ref ,

the intra prediction direction which has been selected if the intra prediction has been applied to the block B _ref during the coding of the latter,

the index of the motion vector if the prediction applied to the block Bre _f is of type inter,

the nullity of the residual of the prediction applied to the block B _ref during the coding of the latter,

- partitioning type applied to the block B _re t when coding the latter,

- the type of transformation chosen,

the value of the quantization step chosen,

the type of filtering applied to the block B _ref , such as, for example, the SAO mode ("Sample Adaptive Offset") used in the HEVC standard,

- etc ....

During the reading C52b), one or more of the aforementioned coding parameters associated with the block B _ref can thus be read.

With reference to FIG. 1A, in C53b), the value of a coding parameter PRC1 read in C52b) is copied for the current block. It is thus advantageously not necessary to code the encoding parameter PRC-i.

According to one embodiment, coding of a syntax element ES_PRCi which indicates whether the coding parameter PRCi of the reference block B _re t is a parameter whose value has been copied or not for the block is coded in C54b). current. The coding C54b) is for example a entropic coding of CABAC type or else an entropy coding of arithmetic type or Huffman type. This coding is implemented by the coding device MC_C of FIG. 2A.

For example, the syntax element ES_PRCi is encoded:

- to the value 1 to indicate that the value of the coding parameter

PRCi was copied for the current block,

the value 0 to indicate that the value of the encoding parameter PRCi has not been copied for the current block and has therefore been classically coded.

In the case where the syntax element ES_PRCi is coded at the value 0, the encoding parameter PRCi is coded in a conventional manner.

According to one embodiment, during the coding C54b), are coded K elements of syntax ES_PRCi, ES_PRC ₂ , ... , ES_P RCK which indicate whether each of the encoding parameters PRC-i, PRC ₂ , ... , P RCK associated with the reference block B _ref is a parameter whose value has been copied or not for the current block.

Of course, it may be decided to code only some of the K syntax elements above. According to one embodiment, assuming that the activateStereoReuse syntax element has been coded in C2 at the value 1, the binary sequence 1 101 means for example that:

- the activateStereoReuse syntax element has been coded in C2 at the value 1,

the value of the type of partitioning that has been applied to the reference block B _ref has been copied for the current block,

the value of the type of transform which has been applied to the reference block B _ref has not been copied for the current block, the type of transform applied to the current block then being coded in a conventional manner,

the value of the quantization step used during the coding of the reference block B _ref has been copied for the current block.

With reference to FIG. 1A, the MCF device of FIG.

C6a) to the construction of a signal portion F which contains, according to the invention, the value 0/1 of said at least one element of syntax ES_PRC- |. The signal portion F also contains all the data of the current block which has been coded in a conventional manner.

The data signal F is then transmitted by a communication network (not shown) to a remote terminal. This includes the decoder DO shown in FIG. 7A.

According to the first embodiment which has just been described with reference to FIG.

the coding operations C1 to C6a) are implemented for each block of the first zone Z1 of the current image IC _j ,

the coding operations C1 to C6b) are implemented for each block of the second zone Z2 of the current image IC _j .

A second embodiment of the coding method according to the invention will now be described with reference to FIG. 1B.

According to this second embodiment of the invention, the encoding method is implemented in a coding device or encoder CO 'shown in Figure 2B which comprises elements similar to those of the CO encoder of Figure 2A. For reasons of simplification, such similar elements are shown in FIG. 2B with the same references as in FIG. 2A.

According to the second embodiment, the first coding method MC1 is applied to any current block located in the first zone Z1, in exactly the same manner as in the first embodiment of FIG. distinguished from that of FIG. 1A in that, with reference to FIG. 1B, in addition to the coding C5b) implemented for the current block located in the second zone Z2, the coding of the block is carried out in C100b) current B _u using the first coding method MC1 applied to any block of the first zone Z1 of the current image, such as for example the coding method shown in FIG. 4. Such a coding 100b) is set implemented by the predictive coding device PRED_C, the block coding device MCB_C and the coding device MC_C which are illustrated in Figure 2B.

With reference to FIG. 1B, the coding methods MC1 and MC2 applied to the current block, respectively in C5b) and in C100b), are put into competition in C200b), according to a predetermined coding performance criterion, for example by minimizing d a distortion rate criterion well known to those skilled in the art.

The competition C200b) is implemented by a computing device CPT as represented in FIG. 2B, which device is controlled by the processor PROC_C.

At the end of the competition C200b), an optimal coding method MC _opt is obtained, such that MC _op t = MC1 or MC _op t = MC2. An element of syntax ES_MC _op t is then coded in C300b).

The coding C300b) is for example a entropic coding CABAC type or even an entropy coding arithmetic type or Huffman. This coding is implemented by the coding device MC_C of FIG. 2B.

For example, the syntax element ES_MC _op t is encoded:

the value 0 to indicate that the current block B _u of the second zone Z2 is coded using the first conventional coding method MC1 selected after the competition C200b),

- to the value 1 to indicate that the current block B _u of the second zone Z2 is encoded using the second coding method MC2 according to the invention, selected after the competition C200b).

With reference to FIG. 1B, the MCF device of FIG.

C400b) to the construction of a signal portion F 'which contains:

if the first conventional coding method MC1 has been selected following the competition C200b):

• the DC _U encoded data obtained in C100b),

· Certain information encoded by the coder CO ', such as for example:

o the type of prediction Inter, Intra, skip or merge applied to the current block B _u , and if necessary, the prediction mode selected, the index of the predictor block obtained _,

o the type of partitioning of the current block B _u if the latter has been partitioned,

the type of transform applied to the data of the current block B _u ,

o etc ....

if the second coding method MC2 according to the invention has been selected following the placing in competition C200b):

"The 0/1 value of the at least one syntax element

ES_PRd,

All the data of the current block which has been coded in a conventional manner.

The signal portion F 'is then transmitted by a communication network (not shown) to a remote terminal. This includes the decoder DO shown in FIG. 7B.

According to the second embodiment of the coding method which has just been described:

the coding operations C1 to C400b) are implemented for each block of the second zone Z2 of the current image IC _j .

Detailed description of the decoding part

A first embodiment of the invention will now be described, in which the decoding method according to the invention is used for decoding a signal or data stream representative of an image or a sequence of images that is suitable. to be decoded by a decoder in accordance with any of the current or future video decoding standards.

In this embodiment, the decoding method according to the invention is for example implemented in a software or hardware way by modifications of such a decoder. The decoding method according to the first embodiment of the invention is represented in the form of an algorithm comprising operations D1 to D7a) or D1 to D7b) as represented in FIG. 6A.

According to this first embodiment, the decoding method according to the invention is implemented in a decoding device or decoder DO shown in FIG. 7A.

As illustrated in FIG. 7A, according to the first embodiment of the invention, the decoder DO comprises a memory MEM_D which itself comprises a buffer memory MT_D, a processor PROC_D controlled by a computer program PG_D which implements the decoding method according to the invention. At initialization, the code instructions of the computer program PG_D are for example loaded into a RAM RAM_D noted, before being executed by the PROC_D processor.

The decoding method shown in Figure 6A applies to any fixed coded current picture IC _j or part of a sequence of L IC-i images, IC _j, ..., IC _L (1≤j≤L ) to decode.

The current image IC _j to be decoded originates from at least one video sequence comprising, as non-exhaustive examples:

images obtained by projection of a mono 360 ° video,

- etc. Optionally, as shown in dotted lines in FIG. 6A, D1 reads, in the data signal F, the coded value 0 or 1 of the activateStereoReuse syntax element which is associated with a characteristic of the current image IC _j to decode. The reading D1 is implemented only in the case where this syntax element has been coded at the level of the current image IC _j .

According to a preferred embodiment, the activateStereoReuse syntax element that is read takes, for example:

the value 1 if the current image to be encoded has been obtained by projection of a stereo video, 360 °, 180 ° or others and that the current image to be decoded has been composed according to the abovementioned FP technique,

the value 0 if the current image to be decoded is of 2D type or has been obtained by projection of a mono video, 360 °, 180 ° or others.

The reading D1 is implemented by a PARS_D device for flow analysis, as shown in FIG. 7A, said device being controlled by the processor PROC_D.

In the case of identification, with reference to FIG. 6A, the decoding of the coded value 0 yes taken by the activateStereoReuse syntax element is carried out at D2.

Such a decoding D2 is implemented by a decoding device

MD_D shown in Figure 7A, which device is controlled by the PROC_D processor.

The decoding is for example a CABAC entropic decoding or an arithmetic or Huffman entropy decoding.

Such a decoding D2 is not necessary in the case where the decoder DO operates independently a distinction between:

a current image to be decoded of 2D type or obtained by projection of a mono video, 360 °, 180 ° or others,

- and a current image to be decoded obtained by projection of a stereo video, 360 °, 180 ° or others and composed seltui a technique of type

FP.

In the remainder of the description, it is considered that the current image to be decoded has been obtained by projection of a stereo video, 360 °, 180 ° or others, and that the current image is composed of several views captured at the same time instant and arranged in the current image so as to form a single view (rectangle of pixels).

With reference to FIG. 6A, the identification of the coded data Dd, DC ₂ ,..., DC _U , DCs (1 u u S S) associated with the blocks B i , B ₂ , B _u , ..., B _s previously encoded according to the above-mentioned lexicographic order, which were obtained at the end of the coding operation C5a) or C5b) of FIG. 1A. that the blocks are decoded one after the other, from left to right, corresponding to the coding order mentioned above.

Such identification D3 is implemented by the PARS_D flow analysis device of FIG. 7A.

Other types of course than the one just described above are of course possible and depend on the order of course chosen coding.

In one example, the blocks B _; B ₂ , B _u , ..., B _s have a square shape and all contain K pixels, with K> 1. By way of non-exhaustive example, the blocks have a size of 64 × 64 pixels and / or 32 × 32 and / or 16 × 16 and / or 8 × 8 pixels.

Depending on the size of the image that is not necessarily a multiple of the size of the blocks, the last blocks on the left and the last blocks on the bottom may not be square. In an alternative embodiment, the blocks may be for example of rectangular size and / or not aligned with each other.

With reference to FIG. 6A, at D4, the decoder DO of FIG. 7A selects as a current block to be decoded a set of coded data current DC _U of the image IC _j , associated with a block B _u to be decoded.

In D5, the current block B _u to be decoded of the image IC _{j is located} , for example by determining its coordinates with respect to the first reconstructed pixel of the image IC _j , located at the top left of this image. last and has coordinates (0,0).

The location D5 is implemented by a computing device CAL1_D as shown in FIG. 7A, which device is controlled by the processor PROC D. If the current block B _u belongs to the first zone Z1 of the image IC _j , with reference to FIG. 6A, the current block is decoded with the aid of a first decoding method MD1 which corresponds to the coding method MC1 applied to the coding, in C5a) in FIG. 1 A. The first method of decoding MD1 is a conventional method, an example of which is illustrated in FIG. 8. The decoding method MD1 is applied to any block current of zone Z1.

With reference to FIG. 8, such a conventional decoding method MD1 implements in D61 a) the decoding of prediction information previously read in the signal F, such as the type of prediction Inter, Intra, skip or Merge applied to current block B _u , and if necessary, the prediction mode selected, the index of the BP _opt- predictor block which was obtained during the prediction C51 a) (fig.4) which has been implemented in the coding.

At the end of the decoding D61 a), the predictor block BP _op t associated with the decoded index is obtained.

The coded data DC _U of the current block B _u are decoded in D62a). Such decoding is implemented by a block decoding device MDB_D shown in FIG. 7A, which device is controlled by the PROC_D processor.

The decoding D62a) implements, in D621 a), a decoding of the DC _U data associated with the current block B _u to be decoded and which have been coded in C5a) in FIG. 1 A. At the end of such a decoding , is obtained a set of digital information associated with the quantized coefficient block Bq _u which has been obtained in C532a) in FIG. 4.

The decoding D621 a) is implemented by the decoding device

MD_D shown in Figure 7A.

The decoding D62a) further implements a dequantization D622a) of the quantized coefficient block Bq _u , according to a conventional dequantization operation which is the inverse operation of the quantization C532a) of FIG. 4. A set of dequantized coefficients current BDq _u is then obtained. Such dequantization is for example of scalar or vector type and is implemented by means of a device MQ ^"1 _D of inverse quantization, as shown in FIG. 7A, which device is controlled by the PROC_D processor.

The decoding D62a) further implements the application D623a) of a transform to the set of dequantized coefficients BDq _u current obtained in D622a). In a manner known per se, such a transform is an inverse transform from that applied to the C531 encoding a) in FIG. 4, such as, for example, a DCT, DST, DWT, LT or other transform. Correspondingly to the CO encoder of FIG. 2A, these transforms are part of a list of transforms LTS1 ^"1 which is previously stored in the buffer MT_D of the decoder DO of FIG 7A The type of transform to be applied can be determined to the decoder conventionally, by reading, in the data signal F, the index of the transform applied to the coding.

The transformation application D623a) is performed by a transform calculation device MTR ^" ¹ _D, as shown in FIG. 7A, which device is controlled by the processor PROC_D.

The inverse quantization MQ device ¹ 1D and the transform calculation MTR device ¹ 1D are contained in a block decoding device MDB_D shown in FIG. 7A, which device is controlled by the PROC_D processor.

At the end of the decoding D62a) of the data of the current block, a current decoded residue block BDr _u is obtained.

With reference to FIG. 8, in D63a), the current decoded residue block BDr _u is added to the BP _opt predictor block obtained in D61 a).

The operation D63a) is implemented by a predictive decoding device PRED_D shown in FIG. 7A, which device is controlled by the PROC_D processor.

A current decoded block BD _U is obtained after the operation D63a).

With reference again to FIG. 6A, the current decoded block BD _U is written in D7a) in a decoded picture ID _j .

The writing D7a) is implemented by an image reconstruction URI device as represented in FIG. 7A, the URI device being controlled by the PROC_D processor. Referring to FIG. 6A, if at the end of the location D5, the current block B _u belongs to the second zone Z2 of the image IC _j , it is carried out at D6b) at the decoding of the current block using a second decoding method MD2 which corresponds to the coding method MC2 applied to the coding, in C5b) in FIG. 1A. According to the first embodiment, the second decoding method MD2 is applied to any current block to be decoded which is located in the second zone Z2.

According to the invention, with reference to FIG. 6A, D61 b) is used to identify a reference block B _ref that has been previously decoded, and which is located in the first zone Z1 of the image. current IC _j being decoded.

The identification D61 b) is implemented by a computing device CAL2_D as shown in FIG. 7A, which device is controlled by the processor PROC_D. The identification D61 b) is identical to the identification C51 b) made at the coding with reference to FIG. 1 A.

According to a preferred embodiment, if the current block which has been located in the second zone Z2 has its first pixel at the top left which has coordinates (x _u , y _u ) in the current image IC _j , then a block of reference B ' _ref is determined in the first zone Z1, as being the block whose first pixel pVef in the upper left has coordinates (x' _ref , y ' _ref ), such that and yVef = y _u - h / 2, where h is the top of the current image IC _j .

Examples of determination of the reference block B ' _ref have already been explained with reference to FIGS. 5A and 5B and will not be described again here.

Since the reference block B _ref has been identified in the first zone Z1, with reference to FIG. 6A, the device PARS_D of FIG. 7A proceeds in D62b) to read, in the signal F, at least one syntax element. ES_PRCi which indicates whether a coding parameter PRCi of the reference block B _ref is a parameter whose value has been copied or not for the current block B _u .

It is then proceeded in D63b) to decode the syntax element

The decoding D63b) is for example a CABAC-type entropic decoding or else an entropy decoding of arithmetic type or of Huffman. This decoding is implemented by the encoding device MD_D of FIG. 7A.

For example :

if the decoded value of the syntax element ES_PRCi is equal to 1, the encoding parameter PRCi is used directly as a decoding parameter PRDi of the current block,

if the decoded value of the syntax element ES_PRCi is equal to 0, the encoding parameter PRCi is then decoded using a conventional decoding method.

According to one embodiment, during coding D63b), are decoded

K elements of syntax ES_PRC _1; ES_PRC ₂ , ..., ES_PRC _K which indicate whether each of the coding parameters PRC-i, PRC ₂ ,..., PRC _K associated with the reference block B _ref is a parameter whose value has been copied or not during the coding of the current block.

Of course, it may be decided to decode only some of the K syntax elements above if only some of the K syntax elements have been encoded at C54b) (Figure 1A). According to one embodiment, assuming that the decoded value of the activateStereoReuse syntax element that has been obtained at D2 has the value 1, the binary sequence 101 means, for example, that:

- the activateStereoReuse syntax element has been coded in C2 at the value 1,

the type of partitioning that has been applied to the reference block B _ref is used directly as a decoding parameter of the current block,

the type of transform which has been applied to the reference block B _ref is not used directly as a decoding parameter of the current block and is decoded in a conventional manner,

the value of the quantization step used during the coding of the reference block B _ref is used directly as a decoding parameter of the current block. With reference to FIG. 7A, at least one decoding parameter PRDi associated with the reference block B _ref is stored in a list LST2 of the buffer MT_D of the decoder DO.

By way of non-exhaustive example, the list LST2 of the decoder DO of FIG. 7A contains a number K of decoding parameters PRD-1, PRD ₂ ,..., PRD _K associated with the reference block B _ref identified and which are respectively identical to the K coding parameters PRC-i, PRC ₂ ,..., PRC _K stored in the list LST2 of the coder CO of FIG. 2A. Examples of such parameters have already been described in the coding method of Figure 1A and will not be described again here.

With reference to FIG. 6A, the value of the coding parameter PRCi associated with the syntax element ES_PRCi for the current block B _{u is} copied to the D64b). For this purpose, the value of the decoding parameter PRDi decoded in D63b) is assigned to the current block B _u .

At the end of the application of the second decoding method MD2 to the current block, a current decoded block BD _U is obtained.

The image reconstruction URI device of FIG. 7A proceeds in D7b) to write the current decoded block BD _U in a decoded picture ID _j .

According to the first embodiment which has just been described with reference to FIG. 6A:

the decoding operations D1 to D7a) are implemented for each block of the first zone Z1 of the current image IC _j ,

the decoding operations D1 to D7b) are implemented for each block of the second zone Z2 of the current image IC _j .

A second embodiment of the decoding method according to the invention will now be described with reference to FIG. 6B.

According to this second embodiment of the invention, the decoding method is implemented in a decoding device or decoder DO 'shown in FIG. 7B which comprises elements similar to those of the decoder DO of FIG. 7A. For reasons of simplification, such similar elements are shown in FIG. 7B with the same references as in FIG. 7A. According to the second embodiment, the first decoding method MD1 is applied to any current block located in the first zone Z1, in exactly the same manner as in the first embodiment of FIG. 6A. The second embodiment differs from that of FIG. 6A in that the second decoding method MD2 is not systematically applied to each current block located in the second zone Z2. For this purpose, once the current block B _u has been located in the second zone Z2 of the current image to be decoded, the syntax element is read in the signal F 'in D100b). ES_MC _op t which indicates which coding method MC1 or MC2 was selected following the competition implemented at the coding in C200b) in FIG. 1B. The reading D100b) is implemented by the device PARS_D of FIG. 7B.

It is then proceeded in D200b) to decode the syntax element ES_MC _op t.

The decoding D200b) is for example an entropy decoding type

CABAC or even an entropy decoding arithmetic type or Huffman. This decoding is implemented by the encoding device MD_D of FIG. 7B.

If the decoded value of the syntax element ES_MC _op t is equal to 1, the current block is decoded using the second decoding method MD2, in exactly the same way as in D5b) in FIG. 6A. according to the first embodiment.

If the decoded value of the syntax element ES_MC _op t is equal to 0, the current block is decoded at D300b) using the first decoding method MD1 which has been applied to each current block of the first zone Z1 of the current image, in exactly the same way as in D5a) in FIG. 6A, according to the first embodiment.

The decoding method MD1 is for example the decoding method represented in FIG. 8. Such decoding 300b) is implemented by the decoding device MD_D, the block decoding device MDB_D and the predictive decoding device PRED_D and which are illustrated in Figure 7B. At the end of the application of the first decoding method MD1 or the second decoding method MD2 to the current block, a current decoded block BD _U is obtained.

The image reconstruction URI device of FIG. 7B proceeds, in D400b), with the writing of the current decoded block BD _U in a decoded image ID _j .

According to the second embodiment which has just been described with reference to FIG. 6B:

the decoding operations D1 to D400b) are implemented for each block of the second zone Z2 of the current image IC _j .

It goes without saying that the embodiments which have been described above have been given for purely indicative and non-limiting purposes, and that many modifications can easily be made by those skilled in the art without departing from the scope. of the invention.

Claims

1. A method of coding an image (ICj) cut into blocks, said image containing distinct first and second zones (Z1, Z2),

said coding method being characterized in that it implements the following, for at least one current block (B _u ) of the image:

determining (C4) to which the first and second zones belong the current block,

if the current block belongs to the first zone (Z1), code

(C5a)) the current block using a first coding method (MC1),

if the current block belongs to the second zone (Z2), coding (C5b) the current block using a second coding method (MC2) which comprises the following:

starting from the position of the current block in the second zone, identifying (C51 b) a block (B _ref ) previously coded, then decoded, situated in the first zone of the image,

- copy the value of at least one encoding parameter (PRC-i) associated with the identified block.

The coding method according to claim 1, wherein the second coding method is applied to all the blocks of the second zone of the current image.

The coding method according to claim 1, implementing the following:

coding (C100b) said at least one current block of the second zone of the image using the first coding method (MC1),

selecting (C200b) the first coding method or the second coding method in accordance with a predetermined coding performance criterion,

coding (C300b) an information (ES_MC _op t) representative of said selection.

The coding method as claimed in claim 1, in which the first and second zones of the image have the same shape, the first zone being situated above the second zone and separated from the latter by a horizontal border extending in the middle of the image.

5. Encoding device (CO) of at least one image (IC _j ) cut into blocks, said image containing first and second distinct zones, characterized in that it comprises a processing circuit (CT_C) which is arranged to implement the following, for at least one current block of the image:

determining which of the first and second zones belongs to the current block,

if the current block belongs to the first zone (Z1), coding the current block using a first coding method (MC1),

if the current block belongs to the second zone (Z2), coding the current block using a second coding method (MC2) which comprises the following:

starting from the position of the current block in the second zone, identifying a block (B _ref ) previously coded and decoded located in the first zone of the image,

- copy the value of at least one coding parameter

(PRC-i) associated with the identified block.

A computer program comprising program code instructions for performing the steps of the encoding method according to any one of claims 1 to 4, when said program is executed on a computer.

A computer-readable recording medium on which a computer program is recorded comprising program code instructions for performing the steps of the encoding method according to any one of claims 1 to 4, when said program is executed by a computer.

8. A method of decoding a data signal (F) representative of an image (ICj) cut into blocks which has been encoded, said at least one image containing first and second distinct areas (Z1, Z2),

said decoding method being characterized in that it implements the following, for at least one current block (B _u ) to be decoded from the image:

determining (D5) to which the first and second zones belong to the current block,

if the current block belongs to the first zone (Z1), decoding (D6a) the current block using a first decoding method (MD1),

if the current block belongs to the second zone (Z2), decode

(D6b)) the current block using a second decoding method which includes the following:

starting from the position of the current block in the second zone, identifying (D61 b) a previously decoded block located in the first zone of the image,

- assign (D64b)) to the current block at least one decoding parameter (PRD-i) associated with said identified block.

9. Decoding method according to claim 8, wherein the second decoding method is applied to all the blocks of the second zone of the current image.

The decoding method according to claim 8, wherein the second decoding method is applied to the current block of the second zone if, for said current block, is read (D1 00b) in the data signal (F '). a selection information of said second decoding method, the first decoding method being applied to the current block of the second zone if, for said current block, is read in the data signal selection information of said first decoding method.

1 1. A decoding method according to any one of claims 8 to 10, wherein the first and second regions of the image have the same shape, first zone being located above the second zone and separated from the latter by a horizontal border extending in the middle of the image.

12. A device for decoding a data signal (F) representative of an image (IC _j ) cut into blocks which has been coded, said at least one image containing first and second distinct zones, characterized in that it comprises a processing circuit (CT_D) which is arranged to implement the following, for at least one current block to be decoded from the image:

determining which of the first and second zones belongs to the current block,

if the current block belongs to the first zone (Z1), decoding the current block using a first decoding method (MD1),

if the current block belongs to the second zone (Z2), decoding the current block using a second decoding method which comprises the following:

assigning to the current block at least one decoding parameter (PRD-ι) associated with said identified block.

Computer program comprising program code instructions for performing the steps of the decoding method according to any one of claims 8 to 11, when said program is executed on a computer.

A computer-readable recording medium on which a computer program is recorded comprising program code instructions for performing the steps of the decoding method according to any one of claims 8 to 11, when said program is run by a computer.